SILBERMAN, Circuit Judge:

In these consolidated cases, petitioners request review of the Interstate Commerce Commission's reversal of a forty-year old policy that substantially prevented railroads from charging the owners of private railcars for the costs of transporting their cars (when empty) to repair depots for ordinary maintenance. We hold that the Commission's reinterpretation of the statute is permissible and rationally explained, and further conclude that it was within the Commission's discretion to apply its new policy retroactively to the parties before it in this adjudication. Accordingly, we deny the petitions in all respects.

I.

Railroads, pursuant to their common-carrier obligations under the Interstate Commerce Act, must provide railcars suitable for the transportation of a broad range of property, including agricultural products, flammable liquids, as well as crated freight. In part because of these diverse requirements, it has proved impracticable for rail common-carriers to invest the capital necessary for the acquisition of general-use and specialty rolling stock. Carriers, therefore, commonly lease railcars both from firms in the business of supplying railcars and, occasionally, from shippers themselves. Under these leasing arrangements, railroads fulfill their common-carrier obligation to make available suitable railcars by paying car providers their costs of owning the rolling stock through a variety of means, including direct "mileage allowances" and offsets on line-haul freight tariffs.

Railcars "provided" to railroads in this manner accrue a substantial amount of "empty mileage" — mileage traveled without carrying any freight — in part because railcars commonly make a so-called "empty return." Railroads are not permitted to charge the railcar owners or car providers separately to return an empty car to the point of origination; instead, carriers may recover the cost of an empty return only by incorporating it in their rates for the linehaul movement of freight. In addition, railcars are periodically moved empty to repair depots selected by car providers. Up until the challenged order was filed, the Commission treated such empty-repair mileage identically to empty-return mileage by prohibiting carriers (in most circumstances) from charging car providers separately for empty-repair moves. In a 1977 adjudication involving many of the parties before us today, see General Am. Transp. Corp. v. Indiana Harbor Belt R.R. Co., 357 I.C.C. 102 (1977) ( "Indiana Harbor I"), aff'd sub nom. Indiana Harbor Belt R.R. Co. v. General Am. Transp. Corp., 577 F.2d 394 (7th Cir. 1978), the Commission reaffirmed its then longstanding prohibition on empty-repair-move charges. It held that private railcars moving empty for repair continued to serve as "instrumentalities of transportation" during such movements since repairs enabled railroads to maintain their use of such cars for revenue-generating purposes. Accordingly, the Commission concluded that empty-repair moves were not "distinct rail services" for which railroads could impose a separate tariff; rather, such moves were deemed a collective responsibility of those carriers who derive more than de minimus economic benefit (through line-haul freight charges) from the use of privately-owned railcars. 357 I.C.C. at 126-27.

Whatever its merits, Indiana Harbor I's regime of collective railroad responsibility for empty-repair moves left certain carriers (predominantly switching or terminating railroads) with an empty-repair mileage burden considerably out of proportion to the economic benefit those carriers derived from the line-haul movement of freight. And cross-subsidization among railroads was not the only perceived evil of Indiana Harbor I: the rule was believed to encourage inefficient depot selections because the costs of empty-repair transportation were imposed on parties other than those selecting the repair depot locations. In 1983, the Indiana Harbor Belt Railroad ("IHB") and the Baltimore Ohio Chicago Terminal Railroad Company ("BOCT"), two carriers that found themselves disadvantaged by Indiana Harbor I, filed proposed tariffs for empty-repair moves in an effort to prompt the Commission to change course. The IHB and the BOCT contended that the statutory basis for the theory of collective responsibility had been undermined by recent congressional enactments designed to give railroads greater flexibility in imposing tariffs for rail services. The proposed tariffs elicited virtually immediate complaints from car providers, and the instant proceeding began.

After an initial dismissal of the complaints by an administrative law judge and a subsequent reversal by the ICC Review Board (holding that Indiana Harbor I controlled the case), the complaints were appealed to the full Commission in July 1984. The ICC stayed the Review Board's order and published a notice in the Federal Register indicating that it was considering modifying or reversing Indiana Harbor I. See 49 Fed.Reg. 39, 740 (1984); see also 50 Fed.Reg. 10,867 (1985) (soliciting additional comments). The Commission thereafter reopened the record underlying the complaints before it (and in Indiana Harbor I) and invited industry comments concerning, among other issues, whether or not the prohibition against empty-repair-move charges was efficient and equitable.

Three years later, after compiling an extensive record, the Commission reversed Indiana Harbor I and ruled that carriers may impose tariffs on car providers for empty-repair moves. See General Am. Transp. Corp. v. Indiana Harbor Belt R.R. Co., 3 I.C.C.2d 599 (1987) ( "Indiana Harbor II"). The Commission revisited the statutory questions analyzed in Indiana Harbor I and concluded that the Interstate Commerce Act, at least after its amendment by the Railroad Revitalization and Regulatory Reform Act ("the 4R Act") and the Staggers Act does not prohibit carriers from imposing an initial charge on car providers for empty-repair moves. See Indiana Harbor II at 606-11. The ICC further explained that permitting carriers to impose initial repair-move charges would redress the problem of intercarrier misallocation of empty-repair-mileage burdens and improve efficiency in the selection of repair depots by placing empty-repair-transportation costs, in the first instance, on car providers. See id. at 613-16. The ultimate responsibility for the costs of owning railcars, including the costs of empty-repair moves, would continue to be transferred to railroads through customary channels — such as mileage allowances. See id. at 614. Since the ICC concluded that Indiana Harbor I's prohibition on empty-repair-move charges no longer was viable under the Act, it upheld the empty-repair tariffs in the adjudication before it. See id. at 620.

The Commission subsequently denied petitioners' request for reconsideration of its Indiana Harbor II ruling, setting the stage for this proceeding. Before us, petitioners argue that Indiana Harbor I's prohibition of empty-repair charges is specifically required under the Interstate Commerce Act; that Indiana Harbor II rests on an unreasonable construction of the Act and is not supported by a consideration of the legally relevant policy considerations; and that it was unlawful for the Commission to apply Indiana Harbor II retroactively to the specific empty-repair tariffs challenged in the underlying adjudication.

II.

We are initially faced with the argument that the Commission's decision fails the first prong of the Chevron test that governs federal judicial review of agency statutory interpretation. See Chevron U.S.A. Inc. v. Natural Resources Defense Council, 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). The Commission's decision, we are told, runs counter to a specific legislative intent barring carriers from charging for the empty movement of privately-owned railcars for repair. We think this contention is insubstantial: neither the statutory language nor its legislative history clearly addresses the point at issue.

Petitioners contend that the Act specifically prohibits carriers from assessing charges for the movement of instrumentalities of transportation not carrying property, and that Congress intended railcars moving empty for repair to be treated as instrumentalities. Petitioners cite 49 U.S.C. §§ 10761(a) and 10102(26) to support their argument, but we find their appeal to those sections of the Act misdirected. Taken together or apart, these passages do not address the issue we confront. Section 10761(a), as petitioners recognize, merely imposes a tariff publication requirement on carriers offering transportation or services within the jurisdiction of the Commission, and in no way purports to restrict categorically the types of services for which tariffs may be imposed. And section 10102(26)'s sweeping definition of transportation plainly hurts, rather than helps, petitioners' strained construction.

Petitioners also rely on 49 U.S.C. § 10728, the provision that permits carriers to "establish separate rates for distinct rail services to encourage competition[,] promote increased reinvestment by rail carriers [, and] encourage . . . increased nonrailroad investment in the production of rail services." 49 U.S.C. § 10728(a) (1982). The best that can be said for this provision, however, is that it begs the question of whether or not an empty-repair move constitutes one such distinct service. To be sure, prior to the Commission's reversal in Indiana Harbor II, the Seventh Circuit twice approved the Commission's former position that empty-repair moves were not distinct rail services, see Indiana Harbor Belt R.R. Co. v. General Am. Transp. Corp., 577 F.2d 394 (7th Cir. 1978) (affirming Indiana Harbor I on direct review); see also Evans Prods. Co. v. ICC, 729 F.2d 1107 (7th Cir. 1984) (sustaining Commission's application of Indiana Harbor I to parties before the court), but neither decision rested on a construction of the Interstate Commerce Act, much less a finding of specific congressional intent. Rather, as we read the opinions, the court simply held that the Indiana Harbor I policy was not arbitrary and capricious. These prior judicial affirmations of Indiana Harbor I do not preclude the Commission from changing course now.

If specific intent is not found in the original legislation, petitioners argue, it was manifested when Congress legislatively ratified Indiana Harbor I. We find that contention equally unconvincing. Congress revisited the Interstate Commerce Act on two occasions subsequent to Indiana Harbor I — in 1978 when it recodified the Act and in 1980 when it approved the Staggers Act. But we see no indication that Congress was either aware of or had approved the Commission's empty-repair-move policy when it recodified the "distinct rail services" provision in 1978 or in 1980 when it amended other portions of the Act. We have held that application of the legislative reenactment doctrine requires a showing of both congressional awareness and "express congressional approval of an administrative interpretation if it is to be viewed as statutorily mandated." AFL-CIO v. Brock, 835 F.2d 912, 915 (D.C. Cir. 1987); see also UAW v. Brock, 816 F.2d 761, 767 (D.C. Cir. 1987); Association of American Railroads v. ICC, 564 F.2d 486, 493 (D.C. Cir. 1977). Petitioners' argument thus is unavailing.

III.

We turn to what we see as the heart of petitioners' case — that the Commission's empty-repair-move policy announced in Indiana Harbor II rests on an unreasonable construction of the Interstate Commerce Act. In this section, we also address petitioners' challenge to the logic and rationality of the Indiana Harbor II decision, for we think the questions posed — has the Commission adopted an impermissible construction of the Act and is its empty-repair-move policy arbitrary and capricious — are quite similar. Both questions require us to determine whether the Commission, in effecting a reconciliation of competing statutory aims, has rationally considered the factors deemed relevant by the Act. See AFL-CIO v. Brock, 835 F.2d at 917; Natural Resources Defense Council v. EPA, 824 F.2d 1146, 1163 (D.C. Cir. 1987); Rettig v. Pension Benefit Guaranty Corp., 744 F.2d 133, 152 (D.C. Cir. 1984); cf. NLRB v. United Food Comm'l Workers Union, 484 U.S. 112, 108 S.Ct. 413, 421, 98 L.Ed.2d 429 (1987) (stating that courts owe deference to agency interpretation "as long as interpretation is rational and consistent with statute").

While Chevron commands deference to the Commission's interpretation of the Act which it administers, we must "look to many factors, among them being the `thoroughness, validity and consistency of . . . [the] agency's reasoning,'" in determining the precise weight to be assigned the Commission's construction in the case before us. Orloski v. Federal Election Comm'n, 795 F.2d 156, 164 (D.C. Cir. 1986) (quoting Federal Election Commission v. Democratic Senatorial Campaign Comm., 454 U.S. 27, 37, 102 S.Ct. 38, 44, 70 L.Ed.2d 23 (1981)). Here, of course, we are not asked to review an agency policy at the point of its initial adoption; the order under review represents a reversal of an agency interpretation in force more or less consistently since 1947. See Union Tank Car Co., 268 I.C.C. 338 (1947). But that factor does not mean judicial deference is any less appropriate; indeed, Chevron itself involved an agency about-face on a significant question of statutory construction. See Chevron, 467 U.S. at 857-58, 104 S.Ct. at 2789. Under these circumstances, however, the Commission bears the burden of rationally explaining its departure from its previous interpretation. See Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 41-44, 103 S.Ct. 2856, 2865-2867, 77 L.Ed.2d 443 (1983); see also Natural Resources Defense Council v. EPA, 725 F.2d 761, 767 (D.C. Cir. 1984). We think the Commission has discharged this responsibility in Indiana Harbor II.

A.

The Commission thought it appropriate to reexamine its empty-repair-move policy for two principal reasons. First, it believed the Indiana Harbor I rule, under which railroads that participate in some measure in the line-haul carriage of freight were forced to bear collective responsibility for the costs of empty-repair moves, resulted in an inequitable allocation of such costs among the participating carriers, see Indiana Harbor II, 3 I.C.C.2d at 604, and in the inefficient placement of empty-repair transportation costs on parties who did not select the repair depot. See id. at 611, 616. Integral to its conclusion about unfair intercarrier allocation was the Commission's disenchantment with its so-called "equalization rule" as a means to mitigate cross-subsidization among railroads. See id. at 605-06. Second, the Commission observed that the legislative environment in which the empty-repair-move rule was announced in Union Tank and affirmed in Indiana Harbor I — had shifted from acceptance of notions of collective railroad responsibility and cross-subsidization among carriers to congressional disapproval of those same ideas. See id. at 607, 610-11. By the time it decided Indiana Harbor II, the Commission believed it was acting against a background of congressional desire for greater efficiency in the management of the nation's rail transportation system through increased reliance on market forces.

The record, it seems to us, adequately buttresses these conclusions. Evidence showed that the prior rule, which prohibited empty-repair charges if the carrier derived greater than de minimus revenues from the line-haul movement of freight, placed unfair burdens on certain carriers. And despite the expectations it expressed in Indiana Harbor I, see Indiana Harbor I, 357 I.C.C. at 125, the Commission was presented with abundant indications that the equalization rule was fundamentally ineffective in leveling the differential burdens imposed by its empty-repair-charge scheme. Because the rule was designed primarily to address provider responsibility for excess non-revenue mileage, and as "there [was] no necessary relationship between each carrier's share of repair move responsibility in terms of loaded revenue miles and its participation in repair move operations," use of the equalization accounts could not be expected to prevent railroad cross-subsidization. Id. at 606 (emphasis added). The Commission cited the experience of one of the Indiana Harbor II carrier-parties to illustrate its point: between 1978 and 1982 the IHB had performed empty-repair switching services which, if charges had been allowed, would have earned the carrier $856,452. Yet the IHB recovered only $96,263 under the equalization rule for the services in question. See 3 I.C.C.2d at 604. Permitting carriers to charge for empty-repair moves, the Commission thought, would resolve the intercarrier-allocation problem without doing violence to the principle of carrier responsibility for the costs of owning railroad cars, for car providers could simply combine the costs of empty-repair moves with other costs of ownership in charging the railroads mileage allowances. Indiana Harbor II was described by the Commission as, in large measure, a change in the "methods of accounting for the repair cost element of private car owners' expenses." 3 I.C.C.2d at 10.

Recent rail transportation legislation supports the Commission's judgment that Congress intended the Interstate Commerce Act to be administered with an eye toward eliminating inefficient ratemaking practices. The Commission cited various provisions in the 4R and Staggers Acts, not as reflective of Congress' specific intent on the question, but rather as indicative of the tension between Indiana Harbor I's collective-responsibility rationale and general congressional purposes. We think the Commission permissibly discerned an emerging congressional purpose to permit carriers to price their services efficiently in accordance with both the costs properly chargeable to such services and market forces in order to "provide all carriers with a realistic opportunity to earn adequate revenues." 3 I.C.C.2d at 611 (emphasis added). The Commission relied on language in section 101(a) of the Staggers Act stating that "it is the policy of the United States Government to allow, to the maximum extent possible, competition and the demand for services to establish reasonable rates for transportation by rail," 49 U.S.C. § 10101a(1) (1982), to "promote a safe and efficient rail transportation system by allowing rail carriers to earn adequate revenues," id. § 10101a(3), and "to encourage . . . efficient management of railroads and . . . the elimination of noncompensatory rates." Id. § 10101a(10). The Commission found further support for this notion in statutory provisions that appear to disapprove of rates that "do not contribute to going concern value" and "encourage cross-subsidization." 3 I.C.C.2d at 611; see 49 U.S.C. § 10701a(b)(3) ("the Commission shall recognize the policy of this [Act] that rail carriers shall earn adequate revenues"); id. § 10701a(c) ("[a]ny rate for transportation . . . that does not contribute to the going concern value of [the] carrier is presumed to be not reasonable"); id. § 10704(a)(2) (Commission shall enable carriers to earn adequate revenues); id. § 10707a(e)(2) (Commission shall consider the extent to which rates which do not contribute to going concern value can be changed).

Of course, these provisions do not compel the approach adopted by the Commission in Indiana Harbor II. None of the provisions is specifically directed to relations between carriers and car providers. Taken together, though, the passages support the reasonableness of the Commission's construction of the Act. Congress' objective of an efficient railroad rate structure, and its attendant goal of ensuring that individual carriers have the opportunity to earn adequate revenues, at minimum runs counter to Indiana Harbor I's premise of collective responsibility. We think, then, that the Commission has tendered a facially adequate explanation for reversing its course by demonstrating how the factual and legal predicates that underlay Indiana Harbor I were either misconceived originally or have since substantially eroded.

Petitioners and intervenor LOSAC nevertheless attack the internal logic and consistency of the Commission's reasoning. LOSAC argues that the premises of the Commission's "accounting change" and "revenue adequacy" explanations of Indiana Harbor II are fundamentally inconsistent; the former implicitly assumes that as between railroads and car providers generally, Indiana Harbor II will be revenue-neutral, while the latter — focused on increased revenues for railroads — explicitly assumes otherwise. LOSAC misapprehends the Commission's justification, however. The ICC's concerns over "revenue adequacy," driven in large part by changes in congressional policy (evidenced in the Staggers Act), had little to do with carriers generally, as LOSAC assumes; rather, the Commission was concerned with the distributive inequity of saddling certain carriers with a disproportionate share of the empty-repair-move burden. 3 I.C.C.2d at 604. Characterized properly, the Commission's "revenue adequacy" and "accounting change" explanations are complementary, not antagonistic.

Still, LOSAC charges that the Commission's "accounting change" justification by itself does not withstand scrutiny. It is argued that Indiana Harbor II is a good deal more than a simple change in accounting methods; granting railroads permission to charge "initially" for empty-repair moves will enable carriers to shift these costs (or at least some of them) to car providers, in violation of the Act's implicit direction that railroads bear the costs of owning railcars used in fulfillment of their common-carrier obligations. See, e.g., 49 U.S.C. § 11121(a) (1982) (providing that rail carriers "shall furnish safe and adequate car service"); Pennsylvania R.R. Co. v. Puritan Coal Mining Co., 237 U.S. 121, 35 S.Ct. 484, 59 L.Ed. 867 (1915) (same). Car providers, according to LOSAC, are not assured that they will, in turn, be able to pass the costs of empty-repair moves back through to carriers through mileage allowance charges, contract offsets, or otherwise. This is so because in recent years carriers have proved reluctant to update allowances in accordance with increases in costs of ownership, and have imposed arbitrary "caps" on mileage formulas. Under the ICC's recent decision in LO Shippers Action Comm. v. Aberdeen Rockfish R.R. Co., 4 I.C.C.2d 1 (1987), aff'd sub nom. LO Shippers Action Comm. v. ICC, 857 F.2d 802 (D.C. Cir. 1988), moreover, railroads have no obligation under the Interstate Commerce Act to reimburse car providers, through allowances, for their complete costs of ownership. Instead, the statute permits the Commission, in exercising its discretionary power to regulate car allowances, to look beyond historical costs to car providers to such factors as conditions in the railcar market. See 49 U.S.C. § 11122(b) (1982) ("In determining the rate of compensation [to car providers], the Commission shall consider . . . the national level of ownership of each type of freight car, and other factors that affect the adequacy of the national freight car supply."); LO Shippers Action Comm., 4 I.C.C.2d at 10. LOSAC therefore contends that it was flatly wrong for the Commission to assume that empty-repair-move costs would continue to be borne by carriers.

To be sure, the Commission may have been a bit facile in its initial opinion in suggesting that its decision involved only an accounting change that would have little or no impact in determining who — as between carriers and car providers — would bear the burden of the costs involved. In its order on rehearing, however, the Commission conceded that car providers may be unable to recoup completely the costs of empty-repair moves from railroads, just as conditions in the railcar market might prevent providers from recovering on a dollar-for-dollar basis their other costs of ownership. And the Commission reiterated, in any event, that the statute simply does not guarantee providers full cost recovery.

We see nothing irrational in the Commission's approach, as clarified on rehearing, toward accounting for empty-repair move costs. It is quite incorrect to assume, as petitioners do, that the Interstate Commerce Act requires the Commission to treat empty-repair-move costs differently than other costs of ownership. As the Commission observed, it would be anomalous to segregate the empty-repair-move component of ownership costs for separate treatment:

The railroads' obligation to compensate car owners for costs of ownership still exists, but the most plausible statutory basis for discharging that obligation within the regulatory framework of published rates is section 11122, which governs private car compensation generally.

Order on Reconsideration at 2. Private car providers have no statutory right to insulate themselves from market forces generally in fixing allowances to be paid by carriers for the use of railcars. It follows that providers have no right to sequester the empty-repair-move component of ownership costs from market forces either. As the Commission stated in LO Shippers Action Comm., "allowing market based car hire returns, instead of prescribing fixed returns, best accommodates the various factors [the Commission is] required by [section 11122 of the Act] to consider." 4 I.C.C.2d at 14. It seems reasonable, we think, for the Commission to conclude that what is appropriate for ownership costs generally is perfectly appropriate for the discrete ownership costs at issue in this case.

The foregoing analysis, in our opinion, fairly disposes of petitioners' related charge that it was irrational for the Commission to attempt to solve a problem of inter-carrier cost allocation with the solution of an extra-carrier (car provider) charge. The argument assumes, in the first instance, that an extra-carrier charge for empty-repair moves is otherwise illegitimate under the Act, an assumption that the Commission has adequately rebutted. The Commission made clear, moreover, that it was the anomaly of treating empty-repair-move costs differently than other costs of ownership (as a collective responsibility of the railroad industry rather than a cost to be borne in direct proportion to a carrier's use of private railcars in the linehaul carriage of freight) that produced the problem of intercarrier cost misallocation. Petitioners' argument, it seems to us, runs aground on section 11122 of the Act, which plainly contemplates initial railroad charges for freight movements with a "pass-back" of ownership costs to carriers through allowances.

B.

Petitioners also contend that the Commission failed to pay heed to numerous legally relevant considerations in fashioning its empty-repair-move policy in Indiana Harbor II. It is argued that, prior to reversing Indiana Harbor I and allowing carriers to charge for empty-repair moves, the ICC should have undertaken an examination of the techniques by which car providers would be able to recoup empty-repair-move costs from railroads. In other words, because the Commission rested its decision in part on the assumption that car providers could use mileage allowance systems and other measures to recover empty-repair-move costs, the Commission was obliged in the same proceeding to engage in full-scale scrutiny of those recovery systems.

We disagree. We are persuaded that the Commission need not have undertaken the elaborate inquiry into mileage allowances requested by petitioners in the Indiana Harbor II proceeding. Petitioners do not complain of any regulatory obstacle to recovery of empty-repair-move costs from carriers, a concern that might warrant immediate Commission attention. Instead, the car providers cite market forces alone as impeding full recovery. But the Commission has recently decided to allow market forces to influence rates of compensation to car providers, see LO Shippers Action Comm., 4 I.C.C.2d at 13-14, and we see no reason why the Commission should be prevented from giving the market some opportunity to assimilate repair-move costs before entertaining a challenge to the adequacy of mileage allowances under section 11122.

The Commission reasonably determined, moreover, that at least some of the mileage-allowance recovery issues raised by petitioners, such as the propriety of railroad mileage caps, involve the effectiveness of allowances in enabling car providers to recover costs of ownership generally, and do not bear directly or exclusively on repair-move costs. And, the Commission observed that it was presently entertaining caps and allowance questions in other proceedings. Petitioners fail to show why their rights under section 11122 would not be protected in a separate adjudication. We have no legitimate grounds, under these circumstances, to quarrel with the Commission's procedural determination concerning the limited scope of the Indiana Harbor II proceeding. "[A]gencies . . . need not deal in one fell swoop with the entire breadth of a novel development; instead, `reform may address itself to the phase of the problem which seems most acute to the [regulatory] mind.'" National Ass'n of Broadcasters v. FCC, 740 F.2d 1190, 1207 (D.C. Cir. 1984) (quoting Williamson v. Lee Optical Co., 348 U.S. 483, 489, 75 S.Ct. 461, 465, 99 L.Ed. 563 (1955)).

Petitioners also assert that the Commission arbitrarily neglected to consider those long-term contractual relationships between railroads, car providers, and shippers which were negotiated in reliance on Indiana Harbor I. We are told that many car manufacturers lease railcars to shippers who, in turn, "provide" the cars to railroads for the transportation of freight. Under typical agreements, the manufacturer-owner receives a flat-rate lease payment from the lessee while the lessee pockets any mileage allowances paid by railroads; no mention is made in these agreements of empty-repair-move charges, because Indiana Harbor I severely restricted the circumstances under which such charges could be assessed by carriers. Superimposing Indiana Harbor II's regime on these arrangements, it is asserted, will require manufacturer-owners to remit empty-repair-move charges to the railroads without any means of recovering these payments from the car lessees. Similarly, petitioners contend that shippers who have negotiated long-term contract rates with carriers, see 49 U.S.C. § 10713 (1982) (providing authority for, and regulating, contract rates for rail transportation in lieu of published tariffs), which rates presumably incorporate an offset amount reflecting the shippers' cost of providing railcars but do not currently account for the possibility of shipper-borne empty-repair charges, will be harmed by Indiana Harbor II.

The gist of petitioners' objection, it seems to us, is that the Commission's previous policy induced railroads, car providers, and shippers to do business one way, and that these settled business practices cannot be overridden by developments in the Commission's regulatory policies no matter how compelling the statutory justifications. This argument assumes incorrectly, we think, that industry practices may prevent the Commission from implementing even a justified and well-explained departure from previous policy. That is simply not sound administrative law. See Norman v. Baltimore Ohio R.R. Co., 294 U.S. 240, 307-08, 55 S.Ct. 407, 415-16, 79 L.Ed. 885 (1935) ("when contracts deal with a subject matter which lies within the control of the Congress, they have a congenital infirmity"); New England Tel. Tel. Co. v. FCC, 826 F.2d 1101, 1110 (D.C. Cir. 1987); see also J.I. Case Co. v. NLRB, 321 U.S. 332, 337, 64 S.Ct. 576, 580, 88 L.Ed. 762 (1944).

Petitioners' last objection to the reasonableness of the Commission's decision is that the ICC neglected to devote sufficient attention to numerous alternative intercarrier schemes — not requiring the reversal of Indiana Harbor I — that allegedly would cure intercarrier cost allocation problems. The Commission failed, we are told, to discuss the option of "intercarrier pooling," under which railroads would pool a percentage of their line-haul freight revenues for distribution among carriers in proportion to their repair-movement burden. Other options, such as intercarrier litigation over "excess" line-haul receipts and more adequate parent contributions of a portion of their line-haul freight revenues to their switching subsidiaries, were allegedly dispensed with by the Commission without adequate consideration or explanation. This "`artificial narrowing of options,'" according to petitioners, cannot be countenanced. International Ladies Garment Workers Union v. Donovan, 722 F.2d 795, 817 (D.C. Cir. 1983) (quoting Pillai v. Civil Aeronautics Board, 485 F.2d 1018, 1027 (D.C. Cir. 1973)); see State Farm Mut. Auto. Ins. Co., 463 U.S. at 46-48, 103 S.Ct. at 2868-69.

This is not a case, however, where the Commission labored under a misperception that the policy ultimately selected was "the only available alternative." Pillai, 485 F.2d at 1027. The record reveals that the Commission deliberated upon all three "alternatives" advanced by petitioners prior to settling on the option of allowing empty-repair-move charges. See Indiana Harbor II, 3 I.C.C.2d at 611-13. The "intercarrier litigation" alternative was reasonably viewed as "complex, costly, and impractical" given the extreme difficulty of determining the extent to which a particular "overcompensated" carrier has been enriched at the expense of a complaining switching carrier. Id. at 612-13. The "parent-subsidiary contribution" alternative, while it might redress imbalances between certain discrete carriers, was also viewed as ill-suited to the resolution of an industry-wide problem because not all undercompensated switching carriers are owned by overcompensated line-haul carriers. Id. at 612. Finally, the record was deemed inconclusive on the question of the viability of intercarrier pooling of line-haul freight revenues; "little was said by the parties about it" even though the alternative was identified by the Commission in soliciting public comments. Id. at 613. The Commission reasonably believed, nevertheless, that permitting carriers to charge for repair moves was simpler than "establish[ing] a second elaborate system of allocating repair-movement responsibility among individual carriers," as "systems are already in place for handling other private car repair and car supply costs." Id. at 614. No further consideration of the alternatives proffered by the car providers, in our view, was legally required.

We conclude, therefore, that the Commission's empty-repair-move policy is based both on a permissible construction of the Interstate Commerce Act and on reasoned scrutiny of the legally relevant considerations. The only question that remains is whether or not the Commission could lawfully apply its new controlling principle to the parties before it in Indiana Harbor II.

IV.

Petitioners advance two arguments against the lawfulness of "retroactively" sanctioning the repair-move charges filed by the IHB and the BOCT and challenged by shippers in Indiana Harbor II. First, they characterize the Commission's action in Indiana Harbor II as a rule, which, under the Administrative Procedure Act, may be given prospective effect only. Second, petitioners argue that even if Indiana Harbor II is labeled an adjudication, retroactive application of the Commission's new empty-repair-move policy to the parties in this case violated fundamental principles of fairness. We find neither argument persuasive.

We need not dwell at length on petitioners' assertion that Indiana Harbor II was a retroactive rulemaking. The former policy largely prohibiting empty-repair-move charges was articulated in an adjudicatory proceeding, and Indiana Harbor II itself was an adjudication. It seems to us presumptively reasonable that a controlling principle announced in one adjudication may be modified in a subsequent adjudication, and petitioners offer little to question that presumption here. As we have said before, "[a]djudicatory decisions do not harden into `rules' which cannot be altered or reversed except by rulemaking simply because they are longstanding." See Chisholm v. FCC, 538 F.2d 349, 365 (D.C. Cir. 1976). Petitioners conceded at argument that, in selecting a controlling principle to govern the dispute before the Commission in Indiana Harbor II, the Commission was at the very least empowered to apply that principle in the future to similarly situated parties. Indeed, basic tenets of administrative law require the Commission to apply its rules consistently in adjudicatory proceedings. See generally Atchison, T. S.F. Ry. Co. v. Wichita Bd. of Trade, 412 U.S. 800, 807-09, 93 S.Ct. 2367, 2374-76, 37 L.Ed.2d 350 (1973). That Indiana Harbor II established an adjudicatory policy to be applied in the future, as well as to the parties before the Commission, is thus of no moment.

Petitioners point to the Commission's requests for public comments under 5 U.S.C. § 553 and contend that that procedure somehow "converted" Indiana Harbor II into an informal rulemaking, see 50 Fed.Reg. 10,867 (1985); 49 Fed.Reg. 39,740 (1984), thereby rendering retroactive application of the empty-repair-move policy unlawful. Yet petitioners cite no authority for their theory that an adjudication is converted into a rulemaking solely because an agency solicits and entertains the comments of those who have an interest in prospective application of the principle under study, and we have found none. Certainly, the practice of permitting amicus filings in judicial proceedings has never been thought a basis for equating judicial decisions with legislation. So we reject petitioners' novel contention.

Relying on our opinion in Retail, Wholesale and Department Store Union v. NLRB, 466 F.2d 380 (D.C. Cir. 1972), petitioners argue in the alternative that even if Indiana Harbor II is regarded as an adjudication, retroactive application of the new empty-repair policy is contrary to law. Petitioners purport to apply the five-factor balancing test set forth in Retail, Wholesale and insist that the statutory interest in applying the new rule to the parties' dispute in Indiana Harbor II is outweighed by the inequity involved in disappointing the car providers' reasonable expectations in reliance on longstanding Commission precedent. Of course, as the Supreme Court has observed, every administrative case in which a new or modified controlling principle is announced "has a retroactive effect," and such retroactivity is "not necessarily fatal to its validity." SEC v. Chenery Corp., 332 U.S. 194, 203, 67 S.Ct. 1575, 1580, 91 L.Ed. 1995 (1947). And because the Administrative Procedure Act does not countenance agency use of adjudicatory powers to announce rules of prospective effect only, see NLRB v. Wyman-Gordon Co., 394 U.S. 759, 765-66, 89 S.Ct. 1426, 1429-30, 22 L.Ed.2d 709 (1969) (opinion of Fortas, J.); id. at 777, 89 S.Ct. at 1435 (dissenting opinion of Douglas, J.); id. at 780, 89 S.Ct. at 1437 (dissenting opinion of Harlan, J.), it seems clear that the circumstances in which a rule may be announced but not applied in an adjudication are few. In part for this reason, courts have interpreted the Retail, Wholesale test to bar retroactive application of a new rule only when such application would work a "manifest injustice." Clark-Cowlitz Oper. Agency v. FERC, 826 F.2d 1074, 1081 (D.C. Cir. 1987) (en banc).

We do not find the parties' reliance interest in the Indiana Harbor I regime to be so great as to warrant the conclusion that retroactive application of the Indiana Harbor II empty-repair policy works a "manifest injustice." Soon after the IHB and the BOCT filed their empty-repair tariffs but prior to their effective date, the parties sought ICC suspension of the tariffs as unlawful, see 49 U.S.C. § 10707 (1982) (providing Commission suspension authority); Southern Ry. Co. v. ICC, 681 F.2d 29 (1982) (describing suspension power), yet the ICC declined to exercise its interim authority. If this action alone did not suggest to petitioners that Indiana Harbor I was being drawn into question, the Commission's 1984 order staying the Review Board's decision that struck down the tariffs removed all doubt about the Commission's inclination. In that order the Commission indicated that the railroads had a "substantial" chance of prevailing in defense of their empty-repair-move tariffs. Thus, for most of the period for which petitioners claim retroactive application of Indiana Harbor II is unfair, petitioners have been aware of Indiana Harbor I's vulnerability. The parties' reliance interest in Indiana Harbor I must be discounted accordingly.

Turning to the other side of the equation, we (like the Commission) think the statutory interest in having Indiana Harbor II apply retroactively is significant in this case. The record before the Commission provided abundant support for its conclusion that the IHB and the BOCT were receiving insufficient compensation through line-haul freight traffic for their handling of empty-repair movements, a predicament in tension with the Staggers Act's objective of providing carriers an opportunity to earn adequate revenues. Under the circumstances, we believe it was perfectly appropriate for the Commission "to apply the law in effect at the time it render[ed] its decision." Bradley v. Richmond School Bd., 416 U.S. 696, 711, 94 S.Ct. 2006, 2016, 40 L.Ed.2d 476 (1974).

For the reasons stated, the petitions for review are

Denied.

* * * * * *

(A) a locomotive, car, vehicle, motor vehicle, vessel, warehouse, wharf, pier, dock, yard, property, facility, instrumentality, or equipment of any kind related to the movement of passengers or property, or both, regardless of ownership or an agreement concerning use; and

(B) services related to that movement, including receipt, delivery, elevation, transfer in transit, refrigeration, icing, ventilation, storage, handling, and interchange of passengers and property.